Kaiserslautern - Fachbereich Informatik
Refine
Year of publication
- 2018 (18) (remove)
Document Type
- Doctoral Thesis (15)
- Conference Proceeding (2)
- Article (1)
Language
- English (18)
Has Fulltext
- yes (18)
Keywords
- Visualization (3)
- Evaluation (2)
- machine learning (2)
- Algorithmic Differentiation (1)
- Artificial Intelligence (1)
- Association (1)
- Automatische Differentiation (1)
- Bluetooth (1)
- Caching (1)
- Collaboration (1)
- Computational Fluid Dynamics (1)
- Computer Supported Cooperative Work (1)
- Decision Support Systems (1)
- Evolutionary Algorithm (1)
- Geoinformationssystem (1)
- Interaction (1)
- Klassifikation (1)
- Linked Data (1)
- Machine Learning (1)
- Manufacturing (1)
- Menschenmenge (1)
- Multi-Variate Data (1)
- Netzwerk (1)
- Node-Link Diagram (1)
- Numerische Strömungssimulation (1)
- Performance (1)
- Process Data (1)
- Prozessvisualisierung (1)
- SPARQL (1)
- SPARQL query learning (1)
- Semantic Web (1)
- Simulation (1)
- Smartphone (1)
- Smartwatch (1)
- Systemdesign (1)
- Visualisierung (1)
- WiFi (1)
- artificial intelligence (1)
- associations (1)
- classification (1)
- collaborative mobile sensing (1)
- crowd condition estimation (1)
- crowd density estimation (1)
- crowd scanning (1)
- crowd sensing (1)
- data sets (1)
- dataset (1)
- embedding (1)
- end-to-end learning (1)
- evolutionary algorithm (1)
- graph embedding (1)
- linked data (1)
- participatory sensing (1)
- semantic web (1)
- stationary sensing (1)
- wireless signal (1)
Faculty / Organisational entity
Postmortem Analysis of Decayed Online Social Communities: Cascade Pattern Analysis and Prediction
(2018)
Recently, many online social networks, such as MySpace, Orkut, and Friendster, have faced inactivity decay of their members, which contributed to the collapse of these networks. The reasons, mechanics, and prevention mechanisms of such inactivity decay are not fully understood. In this work, we analyze decayed and alive subwebsites from the Stack Exchange platform. The analysis mainly focuses on the inactivity cascades that occur among the members of these communities. We provide measures to understand the decay process and statistical analysis to extract the patterns that accompany the inactivity decay. Additionally, we predict cascade size and cascade virality using machine learning. The results of this work include a statistically significant difference of the decay patterns between the decayed and the alive subwebsites. These patterns are mainly cascade size, cascade virality, cascade duration, and cascade similarity. Additionally, the contributed prediction framework showed satisfactorily prediction results compared to a baseline predictor. Supported by empirical evidence, the main findings of this work are (1) there are significantly different decay patterns in the alive and the decayed subwebsites of the Stack Exchange; (2) the cascade’s node degrees contribute more to the decay process than the cascade’s virality, which indicates that the expert members of the Stack Exchange subwebsites were mainly responsible for the activity or inactivity of the Stack Exchange subwebsites; (3) the Statistics subwebsite is going through decay dynamics that may lead to it becoming fully-decayed; (4) the decay process is not governed by only one network measure, it is better described using multiple measures; (5) decayed subwebsites were originally less resilient to inactivity decay, unlike the alive subwebsites; and (6) network’s structure in the early stages of its evolution dictates the activity/inactivity characteristics of the network.
Education is the Achilles heel of successful resuscitation in cardiac arrest. Therefore, we aim to contribute to the educational efficiency by providing a novel augmented-reality (AR) guided interactive cardiopulmonary resuscitation (CPR) "trainer". For this trainer, a mixed reality smart glass, Microsoft HoloLens, and a CPR manikin covered with pressure sensors were used. To introduce the CPR procedure to a learner, an application with an intractable virtual teacher model was designed. The teaching scenario consists of the two main parts, theory and practice. In the theoretical part, the virtual teacher provides all information about the CPR procedure. Afterward, the user will be asked to perform the CPR cycles in three different stages. In the first two stages, it is aimed to gain the muscle memory with audio and optical feedback system. In the end, the performance of the participant is evaluated by the virtual teacher.
We present a study comparing the effect of real-time wearable feedback with traditional training methods for cardiopulmonary resuscitation (CPR). The aim is to ensure that the students can deliver CPR with the right compression speed and depth. On the wearable side, we test two systems: one based on a combination of visual feedback and tactile information on a smart-watch and one based on visual feedback and audio information on a Google Glass. In a trial with 50 subjects (23 trainee nurses and 27 novices,) we compare those modalities to standard human teaching that is used in nurse training. While a single traditional teaching session tends to improve only the percentage of correct depth, it has less effect on the percentage of effective CPR (depth and speed correct at the same time). By contrast, in a training session with the wearable feedback device, the average percentage of time when CPR is effective improves by up to almost 25%.
The Symbol Grounding Problem (SGP) is one of the first attempts to proposed a hypothesis about mapping abstract concepts and the real world. For example, the concept "ball" can be represented by an object with a round shape (visual modality) and phonemes /b/ /a/ /l/ (audio modality).
This thesis is inspired by the association learning presented in infant development.
Newborns can associate visual and audio modalities of the same concept that are presented at the same time for vocabulary acquisition task.
The goal of this thesis is to develop a novel framework that combines the constraints of the Symbol Grounding Problem and Neural Networks in a simplified scenario of association learning in infants. The first motivation is that the network output can be considered as numerical symbolic features because the attributes of input samples are already embedded. The second motivation is the association between two samples is predefined before training via the same vectorial representation. This thesis proposes to associate two samples and the vectorial representation during training. Two scenarios are considered: sample pair association and sequence pair association.
Three main contributions are presented in this work.
The first contribution is a novel Symbolic Association Model based on two parallel MLPs.
The association task is defined by learning that two instances that represent one concept.
Moreover, a novel training algorithm is defined by matching the output vectors of the MLPs with a statistical distribution for obtaining the relationship between concepts and vectorial representations.
The second contribution is a novel Symbolic Association Model based on two parallel LSTM networks that are trained on weakly labeled sequences.
The definition of association task is extended to learn that two sequences represent the same series of concepts.
This model uses a training algorithm that is similar to MLP-based approach.
The last contribution is a Classless Association.
The association task is defined by learning based on the relationship of two samples that represents the same unknown concept.
In summary, the contributions of this thesis are to extend Artificial Intelligence and Cognitive Computation research with a new constraint that is cognitive motivated. Moreover, two training algorithms with a new constraint are proposed for two cases: single and sequence associations. Besides, a new training rule with no-labels with promising results is proposed.
In recent years, enormous progress has been made in the field of Artificial Intelligence (AI). Especially the introduction of Deep Learning and end-to-end learning, the availability of large datasets and the necessary computational power in form of specialised hardware allowed researchers to build systems with previously unseen performance in areas such as computer vision, machine translation and machine gaming. In parallel, the Semantic Web and its Linked Data movement have published many interlinked RDF datasets, forming the world’s largest, decentralised and publicly available knowledge base.
Despite these scientific successes, all current systems are still narrow AI systems. Each of them is specialised to a specific task and cannot easily be adapted to all other human intelligence tasks, as would be necessary for Artificial General Intelligence (AGI). Furthermore, most of the currently developed systems are not able to learn by making use of freely available knowledge such as provided by the Semantic Web. Autonomous incorporation of new knowledge is however one of the pre-conditions for human-like problem solving.
This work provides a small step towards teaching machines such human-like reasoning on freely available knowledge from the Semantic Web. We investigate how human associations, one of the building blocks of our thinking, can be simulated with Linked Data. The two main results of these investigations are a ground truth dataset of semantic associations and a machine learning algorithm that is able to identify patterns for them in huge knowledge bases.
The ground truth dataset of semantic associations consists of DBpedia entities that are known to be strongly associated by humans. The dataset is published as RDF and can be used for future research.
The developed machine learning algorithm is an evolutionary algorithm that can learn SPARQL queries from a given SPARQL endpoint based on a given list of exemplary source-target entity pairs. The algorithm operates in an end-to-end learning fashion, extracting features in form of graph patterns without the need for human intervention. The learned patterns form a feature space adapted to the given list of examples and can be used to predict target candidates from the SPARQL endpoint for new source nodes. On our semantic association ground truth dataset, our evolutionary graph pattern learner reaches a Recall@10 of > 63 % and an MRR (& MAP) > 43 %, outperforming all baselines. With an achieved Recall@1 of > 34% it even reaches average human top response prediction performance. We also demonstrate how the graph pattern learner can be applied to other interesting areas without modification.
Tables or ranked lists summarize facts about a group of entities in a concise and structured fashion. They are found in all kind of domains and easily comprehensible by humans. Some globally prominent examples of such rankings are the tallest buildings in the World, the richest people in Germany, or most powerful cars. The availability of vast amounts of tables or rankings from open domain allows different ways to explore data. Computing similarity between ranked lists, in order to find those lists where entities are presented in a similar order, carries important analytical insights. This thesis presents a novel query-driven Locality Sensitive Hashing (LSH) method, in order to efficiently find similar top-k rankings for a given input ranking. Experiments show that the proposed method provides a far better performance than inverted-index--based approaches, in particular, it is able to outperform the popular prefix-filtering method. Additionally, an LSH-based probabilistic pruning approach is proposed that optimizes the space utilization of inverted indices, while still maintaining a user-provided recall requirement for the results of the similarity search. Further, this thesis addresses the problem of automatically identifying interesting categorical attributes, in order to explore the entity-centric data by organizing them into meaningful categories. Our approach proposes novel statistical measures, beyond known concepts, like information entropy, in order to capture the distribution of data to train a classifier that can predict which categorical attribute will be perceived suitable by humans for data categorization. We further discuss how the information of useful categories can be applied in PANTHEON and PALEO, two data exploration frameworks developed in our group.
Computational problems that involve dynamic data, such as physics simulations and program development environments, have been an important
subject of study in programming languages. Recent advances in self-adjusting
computation made progress towards achieving efficient incremental computation by providing algorithmic language abstractions to express computations that respond automatically to dynamic changes in their inputs. Selfadjusting programs have been shown to be efficient for a broad range of problems via an explicit programming style, where the programmer uses specific
primitives to identify, create and operate on data that can change over time.
This dissertation presents implicit self-adjusting computation, a type directed technique for translating purely functional programs into self-adjusting
programs. In this implicit approach, the programmer annotates the (toplevel) input types of the programs to be translated. Type inference finds
all other types, and a type-directed translation rewrites the source program
into an explicitly self-adjusting target program. The type system is related to
information-flow type systems and enjoys decidable type inference via constraint solving. We prove that the translation outputs well-typed self-adjusting
programs and preserves the source program’s input-output behavior, guaranteeing that translated programs respond correctly to all changes to their
data. Using a cost semantics, we also prove that the translation preserves the
asymptotic complexity of the source program.
As a second contribution, we present two techniques to facilitate the processing of large and dynamic data in self-adjusting computation. First, we
present a type system for precise dependency tracking that minimizes the
time and space for storing dependency metadata. The type system improves
the scalability of self-adjusting computation by eliminating an important assumption of prior work that can lead to recording spurious dependencies.
We present a type-directed translation algorithm that generates correct selfadjusting programs without relying on this assumption. Second, we show a
probabilistic-chunking technique to further decrease space usage by controlling the fundamental space-time tradeoff in self-adjusting computation.
We implement implicit self-adjusting computation as an extension to Standard ML with compiler and runtime support. Using the compiler, we are able
to incrementalize an interesting set of applications, including standard list
and matrix benchmarks, ray tracer, PageRank, sparse graph connectivity, and
social circle counts. Our experiments show that our compiler incrementalizes existing code with only trivial amounts of annotation, and the resulting
programs bring asymptotic improvements to large datasets from real-world
applications, leading to orders of magnitude speedups in practice.
Mobility has become an integral feature of many wireless networks. Along with this mobility comes the need for location awareness. A prime example for this development are today’s and future transportation systems. They increasingly rely on wireless communications to exchange location and velocity information for a multitude of functions and applications. At the same time, the technological progress facilitates the widespread availability of sophisticated radio technology such as software-defined radios. The result is a variety of new attack vectors threatening the integrity of location information in mobile networks.
Although such attacks can have severe consequences in safety-critical environments such as transportation, the combination of mobility and integrity of spatial information has not received much attention in security research in the past. In this thesis we aim to fill this gap by providing adequate methods to protect the integrity of location and velocity information in the presence of mobility. Based on physical effects of mobility on wireless communications, we develop new methods to securely verify locations, sequences of locations, and velocity information provided by untrusted nodes. The results of our analyses show that mobility can in fact be exploited to provide robust security at low cost.
To further investigate the applicability of our schemes to real-world transportation systems, we have built the OpenSky Network, a sensor network which collects air traffic control communication data for scientific applications. The network uses crowdsourcing and has already achieved coverage in most parts of the world with more than 1000 sensors.
Based on the data provided by the network and measurements with commercial off-the-shelf hardware, we demonstrate the technical feasibility and security of our schemes in the air traffic scenario. Moreover, the experience and data provided by the OpenSky Network allows us to investigate the challenges for our schemes in the real-world air traffic communication environment. We show that our verification methods match all
requirements to help secure the next generation air traffic system.
If gradient based derivative algorithms are used to improve industrial products by reducing their target functions, the derivatives need to be exact.
The last percent of possible improvement, like the efficiency of a turbine, can only be gained if the derivatives are consistent with the solution process that is used in the simulation software.
It is problematic that the development of the simulation software is an ongoing process which leads to the use of approximated derivatives.
If a derivative computation is implemented manually, it will be inconsistent after some time if it is not updated.
This thesis presents a generalized approach which differentiates the whole simulation software with Algorithmic Differentiation (AD), and guarantees a correct and consistent derivative computation after each change to the software.
For this purpose, the variable tagging technique is developed.
The technique checks at run-time if all dependencies, which are used by the derivative algorithms, are correct.
Since it is also necessary to check the correctness of the implementation, a theorem is developed which describes how AD derivatives can be compared.
This theorem is used to develop further methods that can detect and correct errors.
All methods are designed such that they can be applied in real world applications and are used within industrial configurations.
The process described above yields consistent and correct derivatives but the efficiency can still be improved.
This is done by deriving new derivative algorithms.
A fixed-point iterator approach, with a consistent derivation, yields all state of the art algorithms and produces two new algorithms.
These two new algorithms include all implementation details and therefore they produce consistent derivative results.
For detecting hot spots in the application, the state of the art techniques are presented and extended.
The data management is changed such that the performance of the software is affected only marginally when quantities, like the number of input and output variables or the memory consumption, are computed for the detection.
The hot spots can be treated with techniques like checkpointing or preaccumulation.
How these techniques change the time and memory consumption is analyzed and it is shown how they need to be used in selected AD tools.
As a last step, the used AD tools are analyzed in more detail.
The major implementation strategies for operator overloading AD tools are presented and implementation improvements for existing AD tools are discussed.
The discussion focuses on a minimal memory consumption and makes it possible to compare AD tools on a theoretical level.
The new AD tool CoDiPack is based on these findings and its design and concepts are presented.
The improvements and findings in this thesis make it possible, that an automatic, consistent and correct derivative is generated in an efficient way for industrial applications.
Fast Internet content delivery relies on two layers of caches on the request path. Firstly, content delivery networks (CDNs) seek to answer user requests before they traverse slow Internet paths. Secondly, aggregation caches in data centers seek to answer user requests before they traverse slow backend systems. The key challenge in managing these caches is the high variability of object sizes, request patterns, and retrieval latencies. Unfortunately, most existing literature focuses on caching with low (or no) variability in object sizes and ignores the intricacies of data center subsystems.
This thesis seeks to fill this gap with three contributions. First, we design a new caching system, called AdaptSize, that is robust under high object size variability. Second, we derive a method (called Flow-Offline Optimum or FOO) to predict the optimal cache hit ratio under variable object sizes. Third, we design a new caching system, called RobinHood, that exploits variances in retrieval latencies to deliver faster responses to user requests in data centers.
The techniques proposed in this thesis significantly improve the performance of CDN and data center caches. On two production traces from one of the world's largest CDN AdaptSize achieves 30-91% higher hit ratios than widely-used production systems, and 33-46% higher hit ratios than state-of-the-art research systems. Further, AdaptSize reduces the latency by more than 30% at the median, 90-percentile and 99-percentile.
We evaluate the accuracy of our FOO analysis technique on eight different production traces spanning four major Internet companies.
We find that FOO's error is at most 0.3%. Further, FOO reveals that the gap between online policies and OPT is much larger than previously thought: 27% on average, and up to 43% on web application traces.
We evaluate RobinHood with production traces from a major Internet company on a 50-server cluster. We find that RobinHood improves the 99-percentile latency by more than 50% over existing caching systems.
As load imbalances grow, RobinHood's latency improvement can be more than 2x. Further, we show that RobinHood is robust against server failures and adapts to automatic scaling of backend systems.
The results of this thesis demonstrate the power of guiding the design of practical caching policies using mathematical performance models and analysis. These models are general enough to find application in other areas of caching design and future challenges in Internet content delivery.